care this time of FPSCR flags, then continue in hardfloats unitl another
instruction looking for FPSCR register and so on..
Dino
-Messaggio originale-
Da: BALATON Zoltan
Inviato: giovedì 30 aprile 2020 17:36
A: 罗勇刚(Yonggang Luo)
Cc: Richard Henderson ; Dino Papararo
; qemu-devel
floats.
I hope to have added some new ideas to discussion, thank a lot Alex!
Dino
-Messaggio originale-
Da: Alex Bennée
Inviato: mercoledì 29 aprile 2020 13:57
A: Dino Papararo
Cc: luoyongg...@gmail.com; BALATON Zoltan ; Mark
Cave-Ayland ; Programmingkid
; Howard Spoelstra ;
qemu-.
Typo correction 😊
" if I take a simple fadd f0,f1,f2 I'll copy value derived from adding f1+f2
into f0 register"
-Messaggio originale-
Da: Qemu-ppc Per conto di Dino
Papararo
Inviato: mercoledì 29 aprile 2020 12:18
A: Alex Bennée ; luoyongg...@gmail.com; BALATON Zolt
econd case.
Could this be a fast solution to start implement hard floats for PPC??
A little of documentation here:
http://mirror.informatimago.com/next/developer.apple.com/documentation/mac/PPCNumerics/PPCNumerics-154.html
Regards,
Dino Papararo
-Messaggio originale-
Da: Qemu-devel Per cont
of results/flags and letting them
choose.
Of course I understand, and in part agree, on all your objections.
Simply I prefer have always a choice.
Best Regards,
Dino Papararo
-Messaggio originale-
Da: Aleksandar Markovic
Inviato: mercoledì 26 febbraio 2020 18:27
A: G 3
Cc: Alex Ben
t hardware.
Regards,
Dino Papararo
Da: Qemu-ppc Per conto di
luigi burdo
Inviato: mercoledì 26 febbraio 2020 14:01
A: BALATON Zoltan ; Programmingkid
Cc: David Gibson ; qemu-...@nongnu.org; qemu-devel
qemu-devel ; Howard Spoelstra
Oggetto: R: [RFC PATCH v2] target/ppc: Enable hardfloat for P
}
In this patch there are a lot of functions can benefit by unrolling loops, with
a huge speed improvement.
Maybe compiler could do it itself but aren't humans still better? 😊
Best Regards,
Dino Papararo
-Messaggio originale-
Da: Qemu-devel Per conto di
Mark Cave-Ayland
Invia
